Team Size, Research Variety, and Research Performance : Do Coauthors’ Coauthors Matter?

Despite team-based works having been dominant in scientific production, little is known about the relationships between team size and research performance in cross-domain research contexts. Drawn from a collaboration network of 114,577 scholars, we first investigate the relationships between team size and both research productivity and impact. In the first-degree network (i.e., a network that consists of the focal author, the coauthors, and the connections among the nodes), productivity is positively associated with team size, while team size and research impact exhibit an inverted-U shaped relationship. In the second-degree network (i.e., a network that consists of the focal author, the coauthors, the coauthors’ coauthors, and the connections among the nodes), both research productivity and impact suggest an inverted-U shaped relationship with the team size of second-degree connections. We also find that research variety exerts a moderating effect on the relationships between team size and research performance. With the increase in research variety, the curvilinear relationship between first-degree team size and research productivity steepens, whereas the inverted-U shaped relationship between first-degree team size and research impact first flattens and then displays a U-shaped relationship. In the second-degree network, team size exerts little influence on research productivity when research variety is extremely small, and such an effect becomes inverted-U shaped when research variety increases. The turning point for the inverted-U shaped relationship between second-degree team size and research impact continues to shift to the left and down as research variety increases.